There are two major dietary sources of ?3 polyunsaturated fatty acid (PUFA): the botanic ?3 PUFA [1-linolenic acid (1LNA, 18:3n-3)] and the marine I3 PUFA [eicosapentaenoic acid (EPA, 20:5n-3) and docosahexaenoic acid (DHA, 22:6n-3)]. We have previously demonstrated that prostate cancer risk can be favorably modified by marine I3 PUFA in a genetically predisposed mouse population and that modulation of prostate cancer development by PUFA is mediated in part through Bad-dependent apoptosis. Our recent evidence shows that active Akt protein localizes on the plasma membrane in prostate tumor tissues from mice on I6 diet, whereas it is primarily localized in the cytoplasm in tumor tissues from mice on ?3 diet enriched with DHA. It is unclear how ?3 fatty acids alter Akt localization, and how this in turn affects the function of Akt substrates such as the proapoptotic Bad protein. In mammals, the first carbon (sn-1 position) of the glycerol backbone in phosphatidylinositols (PIP) is usually linked to a saturated fatty acid, the second carbon (sn-2 position) to an ? 6 PUFA, and the last carbon to inositol. We found that DHA can replace the ?6 fatty acid at the sn-2 position of PIPs in cell culture, thereby generating different species of PIPs. We hypothesize that ?6- and ? 3-containing PIPs will localize to different cellular membranes due to their differences in number and/or position of double bonds, resulting in Akt recruitment and activation at different cellular locations, and consequently differential Bad phosphorylation, apoptosis and tumor suppression. To test our hypothesis, we propose to (1) Determine if different I3 fatty acids are similarly incorporated on the sn-2 position of PIPs and if this incorporation affects their subcellular localization in vitro, (2) Determine if ?3 fatty acid incorporation affects the localization and activity of Akt and Bad in vitro, and (3) Confirm the ability of different I3 fatty acids to alter PIP and Akt localization as well as Bad phosphorylation and to suppress prostate cancer in vivo.